Xinyang Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Xinyang Wang has a total of 61 co-authors. Here are the top co-authors by number of shared publications:
- Qiao Sun
School of Information and Technology (School of Artificial Intelligence), Beijing Forestry University, Beijing, China
4 shared publications - Chen Guo
School of Electronics and Information Engineering, Jinggangshan University, No. 28 Xueyuan Road, Qingyuan District, Ji’an 343009, China
2 shared publications Fu Xu
Beijing Forestry University
2 shared publications- Fu Xu
Beijing Forestry University
2 shared publications - Weiwei Lin
South China University of Technology
2 shared publications - Xiaohui Cui
College of Information Science and Technology, Beijing Forestry University, Beijing 100083, China
2 shared publications
Latest Publications
- Open Access
Exploring Text-Guided Wildlife Counting in Aerial Imagery Without Additional Training
Reliable wildlife counts support population assessment and conservation planning, while aerial imagery enables surveys over large or inaccessible areas. Most automated counting systems are developed for predefined animal categories and require retraining when targets change. Text-guided zero-shot counting offers a flexible alternative by specifying the target through language without category-specific retraining, but how such methods transfer to aerial wildlife imagery remains underexplored. Existing aerial animal datasets vary widely in tasks, categories, environments, resolution, and acquisition conditions, making consistent evaluation difficult. We therefore construct WZSC-5K by consolidating, cleaning, and harmonizing existing resources into a standardized 5000-image benchmark covering 28 animal categories. Existing zero-shot counters developed on natural images face challenges on the aerial wildlife imagery represented in WZSC-5K. Small targets and cluttered backgrounds complicate spatial localization, while top-down views can weaken the correspondence between generic text concepts and animal appearance. We therefore develop a training-free adaptation that uses spatial correspondence from a frozen vision foundation model to improve spatial coherence and incorporates aerial context to improve visual-text alignment. Without additional aerial-domain training, the proposed method reduces MAE and RMSE by 13.82% and 5.35% relative to the baseline model, demonstrating improved transfer under the standardized WZSC-5K evaluation setting.
Open AccessExploring Text-Guided Wildlife Counting in Aerial Imagery Without Additional Training
Reliable wildlife counts support population assessment and conservation planning, while aerial imagery enables surveys over large or inaccessible areas. Most automated counting systems are developed for predefined animal categories and require retraining when targets change. Text-guided zero-shot counting offers a flexible alternative by specifying the target through language without category-specific retraining, but how such methods transfer to aerial wildlife imagery remains underexplored. Existing aerial animal datasets vary widely in tasks, categories, environments, resolution, and acquisition conditions, making consistent evaluation difficult. We therefore construct WZSC-5K by consolidating, cleaning, and harmonizing existing resources into a standardized 5000-image benchmark covering 28 animal categories. Existing zero-shot counters developed on natural images face challenges on the aerial wildlife imagery represented in WZSC-5K. Small targets and cluttered backgrounds complicate spatial localization, while top-down views can weaken the correspondence between generic text concepts and animal appearance. We therefore develop a training-free adaptation that uses spatial correspondence from a frozen vision foundation model to improve spatial coherence and incorporates aerial context to improve visual-text alignment. Without additional aerial-domain training, the proposed method reduces MAE and RMSE by 13.82% and 5.35% relative to the baseline model, demonstrating improved transfer under the standardized WZSC-5K evaluation setting.
- 18 September 2026
- Published by Taylor & Francis in Journal of the Chinese Institute of Engineers
- p. 1-13
The non-inclusive g -extra diagnosability of the exchanged hypercube under the PMC and MM* models
1. Advances in computing and integrated-circuit technology have enabled multiprocessor systems to expand to millions of nodes. In multiprocessor systems, the interconnection network provides the un...
- 18 September 2026
- Published by Taylor & Francis in Journal of the Chinese Institute of Engineers
- p. 1-13
The non-inclusive g -extra diagnosability of the exchanged hypercube under the PMC and MM* models
1. Advances in computing and integrated-circuit technology have enabled multiprocessor systems to expand to millions of nodes. In multiprocessor systems, the interconnection network provides the un...
- Open Access
- 2 September 2026
- Published by Springer Nature in npj Heritage Science
An improved Inception Transformer with Bi-Level Routing Attention and multi-scale feature collaborative enhancement for Chinese ancient architecture image classification
Chinese ancient architecture embodies China’s cultural heritage but remains challenging to classify because of spatially dispersed key components, high inter-class similarity, and multi-scale visual characteristics. To address these challenges, we propose BDM-ViT, an improved Vision Transformer model based on the Inception Transformer. Bi-Level Routing Attention enhances the long-range semantic interactions among highly relevant semantic regions through region-level sparse routing, while a Dual-Dimension Feature Collaborative Enhancement module jointly refines channel and spatial features. A Multi-Kernel Feed-Forward Network captures fine-grained textures and global structurals, and a Joint Discriminative Loss improves classification through boundary optimization, hard-sample learning, and intra-class feature constraints. We also construct the Chinese Ancient Architecture Dataset(CAAD), containing 18 architectural categories. Comparative experiments show that BDM-ViT significantly outperforms state-of-the-art convolutional neural networks and Vision Transformer models, achieving 95.46% accuracy with superior precision, recall, and F1-score. Ablation experiments and visualization analyses further verify the effectiveness and complementarity of each proposed module.
- 2 September 2026
- Published by Springer Nature in npj Heritage Science
Open AccessAn improved Inception Transformer with Bi-Level Routing Attention and multi-scale feature collaborative enhancement for Chinese ancient architecture image classification
Chinese ancient architecture embodies China’s cultural heritage but remains challenging to classify because of spatially dispersed key components, high inter-class similarity, and multi-scale visual characteristics. To address these challenges, we propose BDM-ViT, an improved Vision Transformer model based on the Inception Transformer. Bi-Level Routing Attention enhances the long-range semantic interactions among highly relevant semantic regions through region-level sparse routing, while a Dual-Dimension Feature Collaborative Enhancement module jointly refines channel and spatial features. A Multi-Kernel Feed-Forward Network captures fine-grained textures and global structurals, and a Joint Discriminative Loss improves classification through boundary optimization, hard-sample learning, and intra-class feature constraints. We also construct the Chinese Ancient Architecture Dataset(CAAD), containing 18 architectural categories. Comparative experiments show that BDM-ViT significantly outperforms state-of-the-art convolutional neural networks and Vision Transformer models, achieving 95.46% accuracy with superior precision, recall, and F1-score. Ablation experiments and visualization analyses further verify the effectiveness and complementarity of each proposed module.
- Open Access
- 6 August 2026
- Published by MDPI AG in Electronics
- Vol. 15 (15) , 3483
Causal Benefit-Aware Recommendation for Personalized Learning-Path Features: A Targeting-Policy Framework with Provable Guarantees and Randomized Evaluation
Educational platforms increasingly personalize which AI learning-path features (adaptive homework, learner choice) each student receives. The natural correlational baseline ranks students by predicted performance—deliver the feature to those expected to do well—a heuristic that need not identify who actually benefits. We formalize feature recommendation as a causal targeting-policy problem: rank students by the estimated conditional average treatment effect (CATE) of a feature and recommend to the top of the ranking. We prove three results: (i) causal top-CATE targeting maximizes policy value at any budget and weakly dominates predictive (outcome-based) targeting, strictly when the two rankings disagree; (ii) a split-sample doubly robust evaluation of targeting quality is leakage-free (null-exact in finite samples), whereas the naive in-sample version is optimistically biased; and (iii) greedily targeting by CATE traces the optimal cost–benefit (Qini) frontier, with the deployment rule “recommend when .” We validate the method on 17 randomized embedded experiments from the ASSISTments platform. Because the 40 held-out splits re-partition the same students, we do not treat them as independent replicates: we calibrate every headline comparison against a within-experiment permutation null and an experiment-clustered bootstrap. Under that calibrated inference, causal targeting outperforms predictive targeting for adaptive homework at every budget (permutation , the resolution floor of 200 replicates; Holm-corrected ), while for learner choice the same contrast is directionally consistent but not statistically significant (permutation –; clustered ). The direction is stable in both families: no leave-one-experiment-out refit reverses its sign. Predictive targeting is nonetheless the one rule that is reliably worse than the alternatives, because it recommends the feature to high-performing students who benefit least—realized benefit falls monotonically across predicted-performance deciles (from in the lowest to in the highest). Against a fuller baseline suite, causal (CATE) targeting does not beat random, a simple risk-based rule (target low performers), or treating everyone. Indeed, the estimated benefit ranking is close to noise—split-half rank agreement is – and its calibration slope is , far below the ideal of 1—so the gain over predictive targeting comes from avoiding an actively harmful ordering rather than from recovering individual benefit. A fairness analysis shows why this matters: predictive targeting is regressive, concentrating feature access on high-ability students, whereas causal and risk-based targeting reverse that gradient in this corpus; no policy differentiates by neighborhood opportunity zone. Group-conditional policy values, however, are not individually distinguishable from zero once dependence across students and experiments is accounted for; what survives resampling is the allocation itself—predictive targeting directs fewer of its recommendations to low-ability than to high-ability students (95% CI , experiment-clustered)—so we frame the fairness result as improved access, not established equity gains. The actionable finding is therefore narrow and specific: outcome-based targeting systematically mis-allocates learning-path features and should be replaced by some benefit-aware rule; whether that rule needs to be a learned CATE model, rather than a simple risk-based heuristic, is not established by this corpus.- 6 August 2026
- Published by MDPI AG in Electronics
- Vol. 15 (15) , 3483
Open AccessCausal Benefit-Aware Recommendation for Personalized Learning-Path Features: A Targeting-Policy Framework with Provable Guarantees and Randomized Evaluation
Educational platforms increasingly personalize which AI learning-path features (adaptive homework, learner choice) each student receives. The natural correlational baseline ranks students by predicted performance—deliver the feature to those expected to do well—a heuristic that need not identify who actually benefits. We formalize feature recommendation as a causal targeting-policy problem: rank students by the estimated conditional average treatment effect (CATE) of a feature and recommend to the top of the ranking. We prove three results: (i) causal top-CATE targeting maximizes policy value at any budget and weakly dominates predictive (outcome-based) targeting, strictly when the two rankings disagree; (ii) a split-sample doubly robust evaluation of targeting quality is leakage-free (null-exact in finite samples), whereas the naive in-sample version is optimistically biased; and (iii) greedily targeting by CATE traces the optimal cost–benefit (Qini) frontier, with the deployment rule “recommend when .” We validate the method on 17 randomized embedded experiments from the ASSISTments platform. Because the 40 held-out splits re-partition the same students, we do not treat them as independent replicates: we calibrate every headline comparison against a within-experiment permutation null and an experiment-clustered bootstrap. Under that calibrated inference, causal targeting outperforms predictive targeting for adaptive homework at every budget (permutation , the resolution floor of 200 replicates; Holm-corrected ), while for learner choice the same contrast is directionally consistent but not statistically significant (permutation –; clustered ). The direction is stable in both families: no leave-one-experiment-out refit reverses its sign. Predictive targeting is nonetheless the one rule that is reliably worse than the alternatives, because it recommends the feature to high-performing students who benefit least—realized benefit falls monotonically across predicted-performance deciles (from in the lowest to in the highest). Against a fuller baseline suite, causal (CATE) targeting does not beat random, a simple risk-based rule (target low performers), or treating everyone. Indeed, the estimated benefit ranking is close to noise—split-half rank agreement is – and its calibration slope is , far below the ideal of 1—so the gain over predictive targeting comes from avoiding an actively harmful ordering rather than from recovering individual benefit. A fairness analysis shows why this matters: predictive targeting is regressive, concentrating feature access on high-ability students, whereas causal and risk-based targeting reverse that gradient in this corpus; no policy differentiates by neighborhood opportunity zone. Group-conditional policy values, however, are not individually distinguishable from zero once dependence across students and experiments is accounted for; what survives resampling is the allocation itself—predictive targeting directs fewer of its recommendations to low-ability than to high-ability students (95% CI , experiment-clustered)—so we frame the fairness result as improved access, not established equity gains. The actionable finding is therefore narrow and specific: outcome-based targeting systematically mis-allocates learning-path features and should be replaced by some benefit-aware rule; whether that rule needs to be a learned CATE model, rather than a simple risk-based heuristic, is not established by this corpus. - 17 April 2026
- Published by Elsevier in Information Fusion
- Vol. 134
Modality-aware instruction tuning of large language models for multimodal knowledge graph completion
- 17 April 2026
- Published by Elsevier in Information Fusion
- Vol. 134
Modality-aware instruction tuning of large language models for multimodal knowledge graph completion
- Open Access
Amur Tiger Individual Identification Based on the Improved InceptionResNetV2
Accurate and intelligent identification of rare and endangered individuals of flagship wildlife species, such as Amur tiger (Panthera tigris altaica), is crucial for understanding population structure and distribution, thereby facilitating targeted conservation measures. However, many mathematical modeling methods, including deep learning models, often yield unsatisfactory results. This paper proposes an individual recognition method for Amur tigers based on an improved InceptionResNetV2 model. Initially, the YOLOv5 model is employed to automatically detect and segment facial, left stripe, and right stripe areas from images of 107 individual Amur tigers, achieving a high average classification accuracy of 97.3%. By introducing a dropout layer and a dual-attention mechanism, we enhance the InceptionResNetV2 model to better capture the stripe features of individual tigers at various granularities and reduce overfitting during training. Experimental results demonstrate that our model outperforms other classic models, offering optimal recognition accuracy and ideal loss changes. The average recognition accuracy for different body part features is 95.36%, with left stripes achieving a peak accuracy of 99.37%. These results highlight the model’s excellent recognition capabilities. Our research provides a valuable and practical approach to the individual identification of rare and endangered animals, offering significant potential for improving conservation efforts.
Open AccessAmur Tiger Individual Identification Based on the Improved InceptionResNetV2
Accurate and intelligent identification of rare and endangered individuals of flagship wildlife species, such as Amur tiger (Panthera tigris altaica), is crucial for understanding population structure and distribution, thereby facilitating targeted conservation measures. However, many mathematical modeling methods, including deep learning models, often yield unsatisfactory results. This paper proposes an individual recognition method for Amur tigers based on an improved InceptionResNetV2 model. Initially, the YOLOv5 model is employed to automatically detect and segment facial, left stripe, and right stripe areas from images of 107 individual Amur tigers, achieving a high average classification accuracy of 97.3%. By introducing a dropout layer and a dual-attention mechanism, we enhance the InceptionResNetV2 model to better capture the stripe features of individual tigers at various granularities and reduce overfitting during training. Experimental results demonstrate that our model outperforms other classic models, offering optimal recognition accuracy and ideal loss changes. The average recognition accuracy for different body part features is 95.36%, with left stripes achieving a peak accuracy of 99.37%. These results highlight the model’s excellent recognition capabilities. Our research provides a valuable and practical approach to the individual identification of rare and endangered animals, offering significant potential for improving conservation efforts.
- Top Cited PapersOpen Access
A Recognition Method of Ancient Architectures Based on the Improved Inception V3 Model
Traditional ancient architecture is a symbolic product of cultural development and inheritance, with high social and cultural value. An automatic recognition model of ancient building types is one possible application of asymmetric systems, and it will be of great significance to be able to identify ancient building types via machine vision. In the context of Chinese traditional ancient buildings, this paper proposes a recognition method of ancient buildings, based on the improved asymmetric Inception V3 model. Firstly, the improved Inception V3 model adds a dropout layer between the global average pooling layer and the SoftMax classification layer to solve the overfitting problem caused by the small sample size of the ancient building data set. Secondly, migration learning and the ImageNet dataset are integrated into model training, which improves the speed of network training while solving the problems of the small scale of the ancient building dataset and insufficient model training. Thirdly, through ablation experiments, the effects of different data preprocessing methods and different dropout rates on the accuracy of model recognition were compared, to obtain the optimized model parameters. To verify the effectiveness of the model, this paper takes the ancient building dataset that was independently constructed by the South China University of Technology team as the experimental data and compares the recognition effect of the improved Inception V3 model proposed in this paper with several classical models. The experimental results show that when the data preprocessing method is based on filling and the dropout rate is 0.3, the recognition accuracy of the model is the highest; the accuracy rate of identifying ancient buildings using our proposed improved Inception V3 model can reach up to 98.64%. Compared with other classical models, the model accuracy rate has increased by 17.32%, and the average training time has accelerated by 2.29 times, reflecting the advantages of the model proposed in this paper. Finally, the improved Inception V3 model was loaded into the ancient building identification system to prove the practical application value of this research.
Top Cited PapersOpen AccessA Recognition Method of Ancient Architectures Based on the Improved Inception V3 Model
Traditional ancient architecture is a symbolic product of cultural development and inheritance, with high social and cultural value. An automatic recognition model of ancient building types is one possible application of asymmetric systems, and it will be of great significance to be able to identify ancient building types via machine vision. In the context of Chinese traditional ancient buildings, this paper proposes a recognition method of ancient buildings, based on the improved asymmetric Inception V3 model. Firstly, the improved Inception V3 model adds a dropout layer between the global average pooling layer and the SoftMax classification layer to solve the overfitting problem caused by the small sample size of the ancient building data set. Secondly, migration learning and the ImageNet dataset are integrated into model training, which improves the speed of network training while solving the problems of the small scale of the ancient building dataset and insufficient model training. Thirdly, through ablation experiments, the effects of different data preprocessing methods and different dropout rates on the accuracy of model recognition were compared, to obtain the optimized model parameters. To verify the effectiveness of the model, this paper takes the ancient building dataset that was independently constructed by the South China University of Technology team as the experimental data and compares the recognition effect of the improved Inception V3 model proposed in this paper with several classical models. The experimental results show that when the data preprocessing method is based on filling and the dropout rate is 0.3, the recognition accuracy of the model is the highest; the accuracy rate of identifying ancient buildings using our proposed improved Inception V3 model can reach up to 98.64%. Compared with other classical models, the model accuracy rate has increased by 17.32%, and the average training time has accelerated by 2.29 times, reflecting the advantages of the model proposed in this paper. Finally, the improved Inception V3 model was loaded into the ancient building identification system to prove the practical application value of this research.
- Open Access
The g-Good-Neighbor Conditional Diagnosability of Exchanged Crossed Cube under the MM* Model
Diagnosability plays an important role in appraising the reliability and fault tolerance of symmetrical multiprocessor systems. The novel g-good-neighbor conditional diagnosability restrains that every fault-free node contains at least g fault-free neighbors and is suitable for large scale multiprocessor systems, attracting a lot of research attention. The relationships between the g-good-neighbor connectivity and g-good-neighbor diagnosability of graphs under the model are separately studied, but only applicable in regular graphs or just ranges rather than exact values. As a promising network structure, in 2019, Guo et al. obtained that the g-good-neighbor diagnosability of the exchanged crossed cube () under the PMC model is (). We noticed that the exact value of the g-good-neighbor diagnosability of under the model is still to be determined. In this paper, by proving the upper and lower bounds of the g-good-neighbor diagnosability of , for the first time, we derive that the exact value of its g-good-neighbor diagnosability under the model is (), achieving the unity of the g-good-neighbor diagnosability of ECQ(s, t) under both the PMC model and model. Towards the end, simulation experiments are conducted to evaluate the correctness and effectiveness of our conclusion. Our research provides an important supplement to the g-good-neighbor diagnosability of .Open AccessThe g-Good-Neighbor Conditional Diagnosability of Exchanged Crossed Cube under the MM* Model
Diagnosability plays an important role in appraising the reliability and fault tolerance of symmetrical multiprocessor systems. The novel g-good-neighbor conditional diagnosability restrains that every fault-free node contains at least g fault-free neighbors and is suitable for large scale multiprocessor systems, attracting a lot of research attention. The relationships between the g-good-neighbor connectivity and g-good-neighbor diagnosability of graphs under the model are separately studied, but only applicable in regular graphs or just ranges rather than exact values. As a promising network structure, in 2019, Guo et al. obtained that the g-good-neighbor diagnosability of the exchanged crossed cube () under the PMC model is (). We noticed that the exact value of the g-good-neighbor diagnosability of under the model is still to be determined. In this paper, by proving the upper and lower bounds of the g-good-neighbor diagnosability of , for the first time, we derive that the exact value of its g-good-neighbor diagnosability under the model is (), achieving the unity of the g-good-neighbor diagnosability of ECQ(s, t) under both the PMC model and model. Towards the end, simulation experiments are conducted to evaluate the correctness and effectiveness of our conclusion. Our research provides an important supplement to the g-good-neighbor diagnosability of . - Open Access
- 5 November 2022
- Published by MDPI AG in Mathematics
- Vol. 10 (21) , 4134
HSICCR: A Lightweight Scoring Criterion Based on Measuring the Degree of Causality for the Detection of SNP Interactions
Recently, research on detecting SNP interactions has attracted considerable attention, which is of great significance for exploring complex diseases. The formulation of effective swarm intelligence optimization algorithms is a primary resolution to this issue. To achieve this goal, an important problem needs to be solved in advance; that is, designing and selecting lightweight scoring criteria that can be calculated in time and can accurately estimate the degree of association between SNP combinations and disease status. In this study, we propose a high-accuracy scoring criterion (HSIC) by measuring the degree of causality dedicated to assessing the degree. First, we approximate two kinds of dependencies according to the structural equation of the causal relationship between epistasis SNP combination and disease status. Then, inspired by these dependencies, we put forward this scoring criterion that integrates a widely used method of measuring statistical dependencies based on kernel functions (HSIC). However, the computing time complexity of HSIC is , which is too costly to be an integral part of the scoring criterion. Since the sizes of the sample space of the disease status, SNP loci and SNP combination are small enough, we propose an efficient method of computing HSIC for variables with a small sample in time. Eventually, HSIC can be computed in time in practice. Finally, we compared HSIC with five representative high-accuracy scoring criteria that detect SNP interactions for 49 simulation disease models. The experimental results show that the accuracy of our proposed scoring criterion is, overall, state-of-the-art.- 5 November 2022
- Published by MDPI AG in Mathematics
- Vol. 10 (21) , 4134
Open AccessHSICCR: A Lightweight Scoring Criterion Based on Measuring the Degree of Causality for the Detection of SNP Interactions
Recently, research on detecting SNP interactions has attracted considerable attention, which is of great significance for exploring complex diseases. The formulation of effective swarm intelligence optimization algorithms is a primary resolution to this issue. To achieve this goal, an important problem needs to be solved in advance; that is, designing and selecting lightweight scoring criteria that can be calculated in time and can accurately estimate the degree of association between SNP combinations and disease status. In this study, we propose a high-accuracy scoring criterion (HSIC) by measuring the degree of causality dedicated to assessing the degree. First, we approximate two kinds of dependencies according to the structural equation of the causal relationship between epistasis SNP combination and disease status. Then, inspired by these dependencies, we put forward this scoring criterion that integrates a widely used method of measuring statistical dependencies based on kernel functions (HSIC). However, the computing time complexity of HSIC is , which is too costly to be an integral part of the scoring criterion. Since the sizes of the sample space of the disease status, SNP loci and SNP combination are small enough, we propose an efficient method of computing HSIC for variables with a small sample in time. Eventually, HSIC can be computed in time in practice. Finally, we compared HSIC with five representative high-accuracy scoring criteria that detect SNP interactions for 49 simulation disease models. The experimental results show that the accuracy of our proposed scoring criterion is, overall, state-of-the-art. - 14 September 2022
- Published by Springer Nature in Cluster Computing
- Vol. 26 (3) , 2011-2025
HCIndex: a Hilbert-Curve-based clustering index for efficient multi-dimensional queries for cloud storage systems
- 14 September 2022
- Published by Springer Nature in Cluster Computing
- Vol. 26 (3) , 2011-2025
HCIndex: a Hilbert-Curve-based clustering index for efficient multi-dimensional queries for cloud storage systems